Malaria is the most deadly disease in the populous highlands of Kenya. Efforts to prevent and control malaria epidemics have been limited due to our inadequate understanding of malaria transmission mechanisms. The proposed research will use an interdisciplinary approach to examine the mechanisms leading to the resurgence of highland malaria in Kakamega and Kisii districts, western Kenya. Our hypothesis is that malaria epidemics in the highlands arise because land use changes have significantly altered the microhabitats and microclimatic conditions of vector mosquitoes and the malaria parasite, making local malaria transmission possible in previously unsuitable areas. The proposed research has three specific aims. First, current landuse patterns and recent changes will be examined, through field surveys and satellite image analysis, and the impact of land use changes on mosquito microhabitats and microclimatic conditions will be evaluated. Second, experiments will be conducted to determine whether changes in mosquito habitats and microclimatic conditions increase mosquito vectorial capacity by enhancing mosquito survivorship, shortening the development time of the malaria parasite, and increasing mosquito blood feeding frequency. Third, spatially explicit malaria transmission models will be developed and validated, with a particular emphasis on the effects of land use and microclimate changes. A differential-equation compartment model will be developed to predict threshold densities of vectors, below which epidemics cannot occur, and a spatial stochastic model will be developed to predict malaria risks in heterogeneous environments. The generalizability and prediction accuracy of the models will be tested. This project is innovative in 2 aspects. First, this will be the first detailed study of relationships between land use changes and malaria transmission in the African highlands. Second, this will be the first study to model malaria transmission using a spatial epidemiological modeling approach. We anticipate that our results will have broad application to malaria prevention and control in African.